Studying what politicians communicate reveals what they want audiences to know and how they want those audiences to interpret information. What determines the topics that Members of Parliament (MPs) prioritize in their communications with all constituents across their electoral district? Despite the scholarly attention received by party discipline within Canadian legislatures, researchers have paid comparatively little attention to the extent that party discipline extends to communications outside of the legislature, particularly MPs’ interactions with constituents (Godbout, Reference Godbout2020; Malloy, Reference Malloy2023; Marland, Reference Marland2020; Rayment and McCallion, Reference Rayment and McCallion2023). Indeed, prior research has suggested that constituency operations are one of the last bastions of freedom that MPs have from the full grasp of party discipline in Canada (Koop et al., Reference Koop, Bastedo and Blidook2018). If this local freedom remains true, MPs’ own backgrounds and the constituency context should be stronger predictors of how they communicate with their constituents than the MPs’ political partisanship.
To examine the determinants of MP communications, we collected and examined a sample of constituent newsletters created by MPs’ offices between March 2020 and September 2021 that were prepared for all households and businesses in their electoral districts. Constituent newsletters are an excellent proxy for an MP's communication to their constituents because these publicly funded mail-outs are the most reliable, universal and economical way to reach everyone in the district, as opposed to the subset of constituents and non-constituents who engage with MPs in other ways or other forms of publicity that reach a narrow subgroup. These quarterly newsletters, known as householders, inform constituents about Parliament, government, and local representation and are seemingly created at arm's length from the leader's office and party whip. Every household and business in the MP's electoral district receives a householder, making these publicity instruments a rare opportunity to reach the entire geographic constituency at once, at no direct financial cost to the MP.
In setting out to conduct this study, we expected to find mixed evidence of party cohesion. In Britain, MPs’ constituency newsletters are a respite from repeating a party's centralized messaging because party whips have limited influence over the content (Jackson, Reference Jackson2004), and in Canada there is evidence that MPs and party candidates exert more independence in their electoral districts over localized communications than controlling central actors intend (Koop et al., Reference Koop, Bastedo and Blidook2018; Robbins-Kanter, Reference Robbins-Kanter2022; Yates, Reference Yates, Marland and Giasson2022). These newsletters are one of the only opportunities for MPs to emphasize their own representational style to their constituents (Koop and Marland, Reference Koop and Marland2012), and research suggests there was a period of cross-partisan consensus in public opinion at the outset of the COVID-19 pandemic when we conducted our fieldwork (Merkley et al., Reference Merkley, Bridgman, Loewen, Owen, Ruths and Zhilin2020). However, caucus research bureaus—a team of political staff on the parliamentary payroll who provide communications and research support to MPs but who report to the leader's office—supply MPs’ offices with templates and messaging for speeches, newsletters and social media, and they are on the alert for significant divergence from the party line (Marland, Reference Marland2020: chap. 8; Wilson, Reference Wilson2020). Thus, while householders are an opportunity for MPs to express individualism without party oversight, some staff are likely to avail themselves of centrally produced messaging.
Our analysis of householders produced during the 43rd Canadian Parliament finds that partisanship is a much stronger predictor of differences in communications strategies than MPs’ personal backgrounds or the constituency context. This suggests that the constraints of party discipline in Canada have caught up with at least one element of constituency work: local communication. The results contribute to our understanding of constituency representation and political communication in Canada by shedding light on constituent communications, which is an underexplored element of MPs’ local constituency work. Furthermore, the difficulties that we experienced collecting these publicly funded public documents, including the refusal by some MPs and staff to participate, raises questions about the transparency of MPs’ local communications and the study of Parliament.
Messaging and Constituent Newsletters
Politicians are sensitive about their image, and with good reason: public opinion about how they look and present themselves has implications for election outcomes (for example, Mattes et al., Reference Mattes, Spezio, Kim, Todorov, Adolphs and Michael Alvarez2010). Opponents are incentivized to make elected officials look bad (for example, Basan, Reference Basan2009), which contributes to constituents feeling disconnected from their elected representatives (Coleman, Reference Coleman2004). Public appearances present particular challenges for women and racialized politicians, who must navigate a myriad of stereotypes (for example, Sanghvi and Hodges, Reference Sanghvi and Hodges2015). Accordingly, politicians try to control and disseminate information to shape their public image and persona. Political staff frame how their boss appears in controlled communications such as photographs and advertising, which contributes to shaping a mediated persona that, they hope, will convey that the politician is accessible and busy representing constituents (Corner and Pels, Reference Corner and Pels2003). In Canada, the political environment of strict party discipline is a formidable contextual element given that party leaders and their agents encourage message consistency from MPs (Marland, Reference Marland2016; McGrane, Reference McGrane, Marland, Giasson and Esselment2017). Partisans communicating similar information frames as other partisans is both a function of the social and political forces that bind parliamentarians together (Godbout, Reference Godbout2020; Malloy, Reference Malloy2023; Marland, Reference Marland2020; Rayment and McCallion, Reference Rayment and McCallion2023) and the marketing strategy emanating from a leader's office.
A legislator's personal style also affects how they communicate. Fenno (Reference Fenno1977, Reference Fenno1978) established that a “home style” is the sum of a representative's allocation of resources, their self-presentation to constituents, and their explanations for their legislative work. He suggests that legislators communicate with, among others, a geographic constituency and a re-election constituency. The former refers to the legal boundaries within which local elections are conducted, while the latter refers to the individuals and groups who typically vote for them. From the perspective of democratic representation, a legislator's geographic constituency is the most important, but it is the most difficult to reach. Thus, we should expect the tension between appealing to all constituents, versus just the constituents a legislator needs to win re-election, to influence what content appears in a legislator's localized communication.
A variety of factors influence representational styles in legislators’ newsletters. In the United States, women legislators and representatives of poorer electoral districts are more likely to claim credit for accomplishments in these documents (Dolan, Reference Dolan2006; Yiannakis, Reference Yiannakis1982). In Canada, newsletter content is thought to be similar to party messaging in the legislature, online advertising, and fundraising appeals (McGrane, Reference McGrane, Marland, Giasson and Esselment2017: 157), although other research indicates that MPs use the ephemera to project an image of being in touch with their constituency (Koop and Marland, Reference Koop and Marland2012). For many parliamentarians, a motivator for mailing thousands of newsletters is securing voter support, which is why MPs in marginal seats are incentivized to make use of them (Umit, Reference Umit2017) and why the documents sometimes arrive in mailboxes after an election call (for example, Spurr, Reference Spurr2015).
MPs avail themselves of a publicly funded mechanism to distribute newsletters to their constituents. As set out in the Members By-Law, which is established by MPs on the Board of Internal Economy, the House of Commons administration funds the publication of up to four householders annually, which Canada Post delivers at no cost to the MP's office budget (Canada, House of Commons, 2020). These newsletters profile an MP's work in Ottawa and around the electoral district, provide government information and celebrate community happenings (Figure 1). They enable MPs to communicate directly with all constituents, including those who do not follow them online (Wright Allen, Reference Wright Allen2020). Text, photographs and infographics are packaged within a “personalized template” in a professionally designed layout that uses common headings and allocates plenty of space for photographs (Koop and Marland, Reference Koop and Marland2012: 113–15). Householders are distinct from other forms of constituency mail, including the notorious “ten percenter” flyers that were curtailed in 2010 following backlash against the co-ordinated partisan negativity that was mailed to people outside of MPs’ ridings (CTV News, 2010). Ten percenters were recently restyled as “constituency mail” that encompasses flyers, postcards and reply cards that are sent within an MP's riding; study of those mail-outs would be required to establish to what extent MPs amplify messages from householders and/or communicate different information (Evelyn, Reference Evelyn2019).
Figure 1. Householder front page (householder of Liberal MP Kody Blois, Kings–Hants, April 2020)
When the House of Commons was abruptly suspended in March 2020 after a global pandemic was declared, the ability to publish and distribute householders was paused, even though the need for all MPs to communicate with their entire geographic constituency was on overdrive. MPs complained they needed this service in order to “cater the message about the novel coronavirus to the riding's specific needs” (Wright Allen, Reference Wright Allen2020). A special authorization by the Board of Internal Economy allowed them to hire external printing services to print a householder from mid-April to June 30, 2020, including permission to include “COVID-19-related messages from local community, government or not-for-profit organizations” within the documents (Canada, House of Commons, 2020). During this time, constituency casework skyrocketed as MPs and staff assisted constituents in distress and those who needed help accessing emergency government programs (Koop et al., Reference Koop, Blidook and Fuga2020). However, MPs were unable to attend the public gatherings and events that are content fodder for householders.
During this extraordinary period, we might expect that localized constituency matters and co-ordinated party messages were displaced by nonpartisan information about government programs and services. Normal processes of message co-ordination were upended: different orders of government and all members of political parties spoke in unison during the early stages of the pandemic, and multi-billion-dollar government programs sailed through Parliament in a single day with all-party support. Political elites and the general public entered “a unique period of cross-partisan consensus” in political opinion, where “response to the coronavirus [was] not structured by partisanship” (Merkley et al., Reference Merkley, Bridgman, Loewen, Owen, Ruths and Zhilin2020: 311, 316). Examining MPs’ newsletters is an opportunity to examine how they saw their roles, which messages they thought were most important during the crisis, and the tension between individualism and party-wide message consistency.
Representation and Constituent Contact
The outset of the COVID-19 pandemic may have incentivized Canadian MPs to be less partisan and to communicate to a broader audience. Indeed, when Justin Trudeau initiated a snap election in August 2021, he sought to capitalize on broad support for his government's dramatic public health measures and flare-ups of division in the Conservative Party (Pammett and Dornan, Reference Pammett and Dornan2022). However, Canada entered the pandemic in the context of a minority parliament where electoral considerations are more salient and citizens are more likely to selectively receive and process political information in line with their political beliefs and partisanship (Kevins and Soroka, Reference Kevins and Soroka2018). Furthermore, anti-vaccine mandate and anti-lockdown culture wars were emerging, which came to the fore during the federal campaign and arguably climaxed with the Freedom Convoy protests the following year. Thus, the policy-oriented and vote-oriented instincts of legislators may have been in conflict during the crisis.
What determines the kinds of messages that MPs focus on in constituent communications? Elected officials need to be seen spending time and resources in their electoral districts, and they can do so in a newsletter by communicating that their office can assist with matters such as immigration or taxes (Figure 2). They must further decide how to present their personal image: for example, some MPs use personal touches such as their grandmother's favourite cookie recipe, or a photo with their family. Other MPs may project a good fit because their issue positions match those of their constituents. This can include announcing how they voted on a bill or spotlighting private members’ business that they tabled. Some newsletters solicit opinions from constituents with a postage-free mail-back questionnaire, which (ostensibly) guides the position that the MP should take on a political issue (Koop et al., Reference Koop, Bastedo and Blidook2018: 17, 155). Further, MPs can fashion an image as Ottawa insiders by highlighting the work they do on behalf of their constituents in Parliament (Figure 3). By contrast, some MPs might prefer the image of an outsider or maverick by communicating dissent, which in a system of strict party discipline may serve as a valence signal of integrity and trustworthiness for putting constituents’ interests ahead of the party and the MP's career (Campbell et al., Reference Campbell, Cowley, Vivyan and Wagner2016: 106).
Figure 2. Constituent services in householders (householder of NDP MP Heather McPherson, Edmonton Strathcona, November 2020)
Figure 3. Work updates for constituents (householder of Conservative MP Stephanie Kusie, Calgary Midnapore, Summer 2020)
Since whoever sorts through mail delivered to a constituent's home must glance at the first page of a newsletter to decide whether to read or discard it, and the front portion is what other constituents see when the document lingers within a household, the first page is crucial to promoting an MP's overall image. At the top is a nameplate—the recurring masthead that appears in a fixed position each time and provides key information about what the publication is, such as its title (Koop and Marland, Reference Koop and Marland2012: 122–24). An MP trying to portray an “insider” image may include a nameplate image showing the MP standing in front of Parliament or may use their party's main colour for the newsletter colour scheme (for example, block headings or typeface). Someone trying to portray an “outsider” image may include an image in which the MP is standing in front of a representative community backdrop, such as a city skyline, a crop field, or a representative constituency landmark, or they may use neutral/cross-party colours for their colour scheme (Gulati, Reference Gulati2004; Koop and Marland, Reference Koop and Marland2012). The rest of the front page is composed of information items about policy, the constituency, and so forth.
Research Design
Householders are likely to be a good proxy for the broad way in which MPs want to frame their communications with all constituents. We employ a mixed methods design using quantitative and qualitative data to understand these ephemera in order to increase the validity of findings and our ability to comprehend them (Johnson et al., Reference Johnson, Onwuegbuzie and Turner2007). We collected householders issued by 145 different MPs in the 43rd Parliament, which entailed searching their websites and placing a structured series of requests to their offices. We subsequently obtained some information about the creation of householders by communicating with a convenience sample of MPs and MPs’ staff from a dozen offices.
The quantitative phase involved the collection of householders dated March 2020 to August 2021. Our initial efforts began in January 2021 and were abruptly halted with the snap election call that August. We conducted a second round from August to October 2022 with MPs who were re-elected in the October 2021 federal election. This long data collection period reflects the fact that despite being public documents funded by Parliament, householders can be surprisingly difficult to collect. As there is no central repository, it is up to each MP's office to decide whether to make copies available beyond a one-time mail-out to local households (Koop and Marland, Reference Koop and Marland2012: 121). We anticipated a low response rate given inconsistent work arrangements and other competing priorities during the pandemic, as well as the possibility that some offices might worry that participating would get them in trouble, such as if we were checking to ensure that Board of Internal Economy rules were followed. Election readiness and communications vigilance was noticeable in the lead-up to a rumoured early election call.
We sought out householders from the 334 MPs who were affiliated with a political party at the start of the 43rd Parliament, excluding the Speaker. Five parties were represented in the House of Commons during the 43rd Parliament, which sat from December 5, 2019, to August 15, 2021: the governing Liberal Party, the official opposition Conservative Party, the Bloc Québécois, the New Democratic Party (NDP) and the Green Party. Initially, two research assistants examined MPs’ websites and then emailed the parliamentary offices of remaining MPs to request that recent copies be supplied. In June 2021, a primary investigator emailed the Parliament Hill offices of remaining MPs, asking for the documents again and informing them that a phone call would follow in the event of non-response. In July 2021, a third research assistant rechecked websites and then telephoned remaining MPs’ constituency offices. We paused data collection in mid-August 2021 due to the election and resumed our efforts a year later. In August 2022, we used Canada Post to mail a personalized letter to each MP's constituency office in their preferred official language, as indicated on the Parliament of Canada website. In October 2022, we placed a final round of phone calls to constituency offices of MPs who had not responded. We are confident that employing these multiple attempts over 11 combined months of fieldwork exhausted all reasonable avenues of collecting these documents.Footnote 1 Future research should consider crowdsourcing as a method of obtaining similar documents (Rea et al., Reference Rea, Kleeman, Zhu, Gilbert and Yue2020).
How MPs’ offices handled our requests varied considerably, with the only observed commonalities being that some staffers refused after consulting with their MP and that Conservatives were more hesitant. A few Conservatives flatly stated that they do not participate in research studies, while some Liberal ministers’ parliamentary offices asked for additional information, with one requesting that we complete a consent form. The pandemic context likely encumbered our data collection because staff working remotely did not have easy access to electronic or printed householders. Some reasons for withholding the documents were cryptic, as with the Conservative staffer who explained that “we are not providing any digital copies due to concerns that were raised regarding editing of the digital proofs.” We infer that some MPs and their staff worried that disclosing their householders to non-constituents risked political damage or embarrassment.
After limiting our sample to householders that were written after the World Health Organization declared a pandemic on March 11, 2020 but before the federal election of September 2021, we were left with householders from 145 MPs—a completion rate of 43 per cent. One householder per MP in the study timeframe is examined. If we collected multiple householders issued by an MP, we examined the first one published after the pandemic was declared among those supplied. This ensured the research team had no discretion to choose between different householders or implicitly select those that conformed to pre-existing expectations. The partisan composition of our sample compared to party standings in the 43rd Parliament is shown in online Appendix 1. Overall, Liberal MPs are slightly overrepresented, Conservative MPs are somewhat underrepresented, and the other MPs are appropriately represented.Footnote 2
Two research assistants (one anglophone, one francophone) coded each householder itemFootnote 3 by noting the topic discussed in the heading and article body, resulting in the identification of 67 broad topics. We recoded this list into five sets of frames: policy, the pandemic, constituency, Ottawa, and leader. The list of original topics, how they were defined, and how they were recoded is available in online Appendix 2. A third (bilingual) research assistant coded 20 per cent of the original coders’ work to check for intercoder reliability. The average reliability score using our five final categories was 83 per cent.Footnote 4 These categories represent our dependent variables.
We categorized articles referring to a specific policy area as pertaining to policy. A wide range of policy topics were identified, including housing, the military, healthcare, infrastructure, agriculture, childcare, the economy, disability, language, issues of concern to particular groups such as women and 2SLGTQI persons, racism and the Black Lives Matter movement, to name some. This category also included private member's bills and motions. While this category included discussions of COVID-19 where applicable, we created a second category pertaining specifically to the pandemic. This included information both on COVID-19 policy (for example, citizen repatriation and financial support) and about how to access applicable government financial support, as well as general information such as messages of hope, about social distancing, or vaccine distribution.
The third category we examined were items focusing on the MP's constituency. This included, for example, information on how to access government services, such as websites that help constituents find employment, apply for government benefits or permits, or obtain supports for families, children, and mental health. A nameplate was coded as constituency-based if it predominantly contained images relating to the MP's electoral district (for example, skyline, buildings, rural scenery, trees/lakes, constituents, a constituency office) or contained images of the MP apparently in the riding (for example, wearing a hard hat, digging a hole for a sapling). Some householders include a feedback form where MPs invite constituents to email or to mail back a section of the newsletter to register their opinions on a certain issue. This includes constituency “referenda,” where an MP puts forward a question and suggests that constituents’ responses will inform how they vote in the House of Commons. Another facet of our constituency grouping was community-focused messages that were positive or folksy. This included congratulatory messages to school graduates, well-wishes to those celebrating a national holiday, and food recipes.
Our fourth category involved coding articles as having an Ottawa focus if the MP elaborated on matters specific to parliamentary procedure or the “Ottawa bubble,” such as prorogation, confidence votes, the auditor general's report, or the WE Charity scandal (see Jefferess, Reference Jefferess2021). Instances of Ottawa-based imagery in the nameplate were coded as having an Ottawa focus, including the Parliament Buildings, Peace Tower, Centennial Flame, a parliamentary office, or standing up in Question Period. Finally, we coded householders as emphasizing the leader if there was a message from, or a profile of, the party leader. These were relatively straightforward to identify given that they were framed explicitly as focusing on the leader, rather than the MP.
Since a single item could pertain to more than one topic, these categories were not mutually exclusive. We coded each of them based on whether the content was present in the householder, whether each topic appeared on the first page, and the number of times a topic appeared. We use these data as the dependent variable (present/not present) in a series of logistic regressions (N = 145).Footnote 5 We divide our independent variables into three groups: MP characteristics, riding characteristics, and the broader context. This is because an MP's communication strategy is likely to be the product of their own internal motivations (or those of their party), what they perceive to be the demands of their constituents, and whether there is a dominant issue to which they want to appear responsive. As we describe below, the specific variables we include are guided by prior research.
In terms of MP characteristics, we collected data from Elections Canada and the Library of Parliament about each MP's political party, vote share in the 2019 federal election, their gender, whether they were a rookie MP, and whether they were in cabinet when the householder was published. We investigate the impact of partisanship because of the strong impact party discipline has on Canadian parliamentary politics, whereby MPs of the same party may be incentivized to adopt similar strategies. We control for MPs’ vote shares because legislators from marginal ridings have been shown to prioritize a constituency focus in their representational activities, which may decrease their ability to focus on other things (Heitshusen et al., Reference Heitshusen, Young and Wood2005). With regard to gender, research has shown that women legislators are more likely than men to prioritize constituency service (Thomas, Reference Thomas1992), which may lead women legislators to stress these elements of their job within the context of constituent communications. This is also the case with more junior MPs, since prior research has shown that their more senior colleagues are less likely to adopt a constituency focus as they contemplate other goals (Umit, Reference Umit2017). Finally, we have conflicting expectations with regard to cabinet responsibilities; on the one hand, prior research suggests ministerial office is related to declining constituency focus (Heitshusen et al., Reference Heitshusen, Young and Wood2005); on the other, cabinet ministers may try to overcompensate for their perceived distance from their constituents by trying to lean harder into presenting an image of local responsiveness.
With regard to riding characteristics, we use data from Statistics Canada on the population density per square kilometre of each riding as an approximate measure of urbanization and on median riding income as a measure of the affluence of a riding. These variables are based on prior research that shows US legislators from more affluent and urbanized electoral districts are more likely to focus on national policy issues in newsletters to their constituents, while legislators from lower-income districts emphasize benefits secured for the local electorate (Yiannakis, Reference Yiannakis1982). We also drew on data from Lucas and Armstrong (Reference Lucas and Armstrong2021), who estimate each riding's average ideological position using 56 policy-related survey items from the 2019 Canadian Election Study. While we expect (and as they demonstrate) a strong correlation between the ideology of a riding and the vote share of a given party in each riding, we use this measure to examine whether riding ideology affects how MPs frame their communications, over and above their own partisanship. Finally, we include a categorical variable corresponding to the time period in which a newsletter was published to represent the broader policy context in which the householder was written. COVID-19 was the overwhelming focus of public policy and national discourse during this time; however, we expect MPs to focus less on pandemic-related issues and more on other issues the further away from the reference period (March to June 2020) a newsletter was written.
As we were conducting our quantitative analysis, we became puzzled about how decisions are made in MPs’ offices to prioritize content in their householders. We proceeded to initiate some information exchanges in early 2023 with five MPs and the political staff of seven other MPs, including four telephone interviews and eight email conversations. We selected these 12 different MPs’ offices because of the opportunity to have frank conversations: some had offered to speak about householders during our quantitative phase, two were our local MPs, and the rest had a working relationship with the authors from prior research projects. They represented two ridings in each of Alberta, British Columbia, Nova Scotia, Ontario and Quebec, and one in Newfoundland and Labrador. Participants were affiliated with the Liberals (six), Conservatives (four) and the NDP (two). Among them were five women and seven men. We anchored our exchanges in the following questions: When constructing a householder, how does an MP's office decide between emphasizing things happening in Parliament or in government versus emphasizing things happening in the electoral district? Do you think constituents read your householders? What is the most useful information in a householder for them? Is there anything else that we should consider about what goes into whether to emphasize national, regional or local information in a householder? We ceased recruiting subjects when we felt we had achieved information saturation. A limitation of this qualitative supplement is that we did not interview any unilingual francophones, notably from the Bloc Québécois. Greens, Independents and former MPs and staff might have offered additional perspectives, as would questions about the themes in our findings.
How Do MPs and Their Staff Approach Newsletters?
Members of Parliament and their staff view householders as a cost effective way to generate awareness of the MP. Householders reach people who do not pay attention to politics in the news or on social media, and they help MPs build a personal brand. As one staff member put it to us, “A householder is in theory our #1 marketing tool. It is the only tool we know that will reach every household in the riding” (Liberal staffer Jennifer MacKinnon). MPs’ offices recognize there is wastage—“Some constituents read them carefully, others will review sections of interest to them, and others will discard them” said another staffer (Conservative 1)—yet they also know that a sizable contingent and cross-section of their constituency looks at a householder. For a week following each mail-out, there is a noticeable increase in phone calls and emails with constituency casework, including requests for celebratory certificates for birthdays and anniversaries, as well as more contact from politically attuned constituents sharing policy opinions and reactions. MPs are also told on doorsteps that constituents have been following updates in the householders. Senior citizens are widely thought to be the biggest audience.
The production of a householder is part of the recurring duties of political staff. They anticipate what the MP will approve, and they use past editions and subheadings as a guide. Some MPs want editorials; others like snippets of news. “Where these mailers are going directly into people's homes, we strive to create a digestible, informational and engaging piece—something that would be easy to read over a morning cup of coffee,” explained one MP's office (Liberal 2). There might be events they want to promote or react to, or amplify issues that constituents want raised in the House, or praise or criticism of a minister if an issue matters locally. Householder content is guided by House of Commons rules. Staff in the printing services division reject a submission if content is deemed excessively partisan or other guidelines are not followed, such as a need for at least two modes of communication for constituents to contact their MP. The MP typically reviews a draft and might suggest changes. Occasionally, they draw inspiration from what other MPs are doing.
An MP's representational style reflects their individual persona. Some want to promote their community engagement, some want to highlight a private member's bill that they are proud of, while others like to pen missives to their constituents. Whatever the topic, photographs are important to attract attention, for audience recall of information, and to fill space. Some MPs are routinely accompanied by a staffer who takes photographs of the MP presenting a local award, flipping pancakes, or handing out lapel pins. Staff like including photos of constituents who have a local profile, as well as lesser-knowns whose inclusion transforms them into local mini-celebrities. MPs and their staff believe that blurbs longer than a few hundred words are less likely to be read, with the exception of an MP's opening letter to constituents. The stage of an MP's career has some bearing on visual content: we heard that as a rookie, one MP prioritized photographs to generate name recognition and to assure constituents that she was spending time in the riding. After multiple re-elections, she felt that it was viable to allocate more space to policy issues.
MPs see a need for a balance of information about Ottawa and the constituency, and they prioritize local relevance. For instance, constituents should be informed about a proposal to redraw electoral boundaries or an announcement that a local bridge will be constructed. National information is highlighted for its local relevance, such as details of tax break entitlements. Staff draw out the relevant aspects of government announcements that they believe will resonate in the riding, informed in part by what constituents are contacting them about. Staff try to make information temporally relevant, such as an MP's overseas trip with a parliamentary group that will be framed in the householder within the context of something topical so that it remains timely months later. It is useful to present solutions to a public policy problem and to avoid delving into provincial and municipal areas unless the MP is directly involved. One MP suggested to us that political and parliamentary content increases as an election approaches and partisanship ramps up.
Householders are a forum for trumpeting government accomplishments. Ministers’ staff sometimes try to put a local spin on the minister's activities, or else they stick to constituency matters in order to avoid conflating the roles. Parliamentary secretaries and government-side backbenchers are inclined to feature more information about government initiatives in an effort to support the party. Some offices field more positive feedback when they present neutral information about local developments, as opposed to spinning government language. Opposition MPs present government information deemed useful to constituents, such as a list of phone numbers, but they also take partisan jabs.
Inviting constituents to share their opinions demonstrates that their MP is listening and is a useful source of public opinion research. Feedback is plentiful if readers are invited to tell their MP what they think about a hot-button topic. Constituents fill out forms and mail them back, and a broader spectrum of people participates than those who otherwise contact the MP's office. One staffer explained:
We recently sent out a long-form questionnaire to a portion of the riding and in the more than 300 responses we got back, people said that the most useful information for them is learning about the MP's parliamentary work in Ottawa. Obviously, thousands of [householders] end up inevitably in the recycling bin, however, part of being an MP is informing constituents about the work of our office. The householders are there to inform the population, but they're also an incredible tool to gauge public interest in an issue as well as an office's mobilization capacity. (NDP 1)
Constituents complain about errors as simple as a typo. Anxieties about making a mistake or producing an unprofessional or controversial document cause some offices to miss deadlines. The fear of backlash contributes to paranoia among some MPs and staff that people harbour vendettas and want to embarrass them or that opponents want to critique the messaging. Consequently, some offices are suspicious when non-constituents request copies of householders. Not all MPs are so concerned; as Conservative MP Dan Albas put it to us, “If it's fit enough to be sent to constituents, it should be good enough for anyone. These things should not be treated as national security secrets.”
The partisan slant of content is galvanized when MPs’ offices avail themselves of support from the caucus research bureau (see Marland, Reference Marland2020: chap. 8; Wilson, Reference Wilson2020). These offices have been involved in creating MP newsletters since at least 1973 (Canadian Press, 1973). An MP's staff may lack the technological skills to assemble a newsletter on their own, which takes more time and results in delays, such as when printing services return the draft because an image resolution is too low. A template with suggested lines supplied by the caucus research bureau save time and grief because it specifies where to insert text and photos, and it offers a professional graphic design. An MP's staff examine the supplied messages to identify what topics will resonate locally and which ones to avoid. For example, news that the government is transferring health care funds to the province could be a touchy topic because constituents are complaining about an urgent need for medical personnel now and want a tangible solution.
In addition, many MPs also issue annual calendars that constituents use and praise, and some Conservatives customize a popular tax guide that summarizes tax changes and tax credits. MPs walk a fine line—if they are too impartial, their loyal supporters raise eyebrows; if they are too partisan, other constituents become annoyed.
Which Factors Matter Most as MPs Communicate through Newsletters?
What were MPs talking to their constituents about during COVID-19? Figure 4 shows the distribution of topics contained in our sample of constituent newsletters, including overall mentions (that is, whether a topic appeared in the newsletter at all), first page mentions and the average number of items that discuss each category of topic.
Figure 4. Topics in constituent newsletters (overall mentioned, first page mentions and average number of mentions)
As expected, Figure 4 shows that constituency-focused content is the bread and butter of MP householders. Every householder (100 per cent) in our sample included articles or nameplates focusing on help for the MP's constituents at least once—indeed, the overwhelming majority (approximately 98 per cent) led with constituency content on the first page. Many of these articles included discussions of things like how to access government services and benefits, such as employment assistance, immigration, and other supports. Constituency-focused content also occurred more than any other type of article throughout the entirety of the newsletter; on average, an MP householder had 6.6 items (articles and nameplates) that were framed as appealing to that MP's constituency—almost double the next most popular frame, which was a focus on the COVID-19 pandemic. It is unsurprising that COVID-19-focused articles were the second most prominent topics in householders during our period of observation, since the pandemic and its considerable impacts was the most salient issue for many Canadians in 2020 and 2021 (Environics Institute, 2022). MPs were keen to tell constituents about the availability of vaccines, to discuss financial supports available during COVID-19 lockdowns, and in the case of the opposition, to criticize the government's handling of the pandemic. This appears to reflect that MPs were interested in presenting themselves as responsive to the main issue of the day, which may have been driven by a desire to convey that they were addressing an issue that was very important to Canadians but also, by necessity, involved a large role for government.
It is worth highlighting here again that our coding frame did not assign topics as mutually exclusive—for example, we coded a discussion of pandemic financial supports as having both a COVID-19 focus and a constituency focus, while being critical of lockdowns was coded as both having a COVID-19 focus and a policy focus. The point of doing so was to add additional context about which supports and which policies MPs were most interested in discussing. In both cases, it is predictable that the pandemic was the lens through which many other discussions took place.
MPs led with COVID-19 content on their first page 84 per cent of the time in our sample, and overall, approximately 92 per cent of householders mentioned the pandemic at least once. In terms of overall mentions and the average number of mentions, the pandemic was roughly tied with a discussion of other policy issues, such as housing, agriculture, the environment, and Indigenous affairs. Despite this, MPs were much more likely to lead with a discussion of COVID-19 than other policy areas, which likely reflects that MPs were trying to capture the attention of their constituents; indeed, our interview data showed MPs believe that some constituents only look at that first page.
Fewer MPs adopted an Ottawa-based frame in their householder content. Less than half (49.7 per cent) of householders included Ottawa-based articles at all, and even fewer MPs included House of Commons and related content on their first page (approximately 42 per cent). This likely reflects the fact that MPs did not think their constituents would be especially interested in what goes on within the “Ottawa bubble,” such as whether Parliament would be prorogued, what was in the auditor general's report, or more general discussions of government accountability. Fewer MPs still focused on their party leader—11 per cent of legislators did this at least once throughout their householder, and around 7 per cent led with the leader on their front page.
Unlike other forms of MP communications, a householder ensures that constituents who check their mailboxes cannot avoid exposure to their MP, since even at a cursory glance, they would process what the document is. Since the front page of the householder is the most likely to be read by constituents, we focus on the subset of messages that appear on this page for the remainder of this article. However, since adopting a constituency frame was the overwhelming norm, rather than the exception, there is not enough variability in this item to include it in our model. Therefore, we examine whether MPs focused on the remaining four frames—COVID-19, policy, Ottawa, or their party leader—with the understanding that the backdrop is a heavy focus on constituency content.
Table 1 shows the results of a logistic regression for a series of binary, focus-related dependent variables (COVID-19 focus, policy focus, Ottawa focus, and leader focus), while Tables 2 and 3 show the predicted probabilities from these models for the two variables that are broadly statistically significant (partisanship and time period, respectively). The results demonstrate the importance of political partisanship in framing an MP's overall communication. All else being equal, Conservative MPs were much less likely to focus on COVID-19 than Liberal MPs (though at p = .066, this coefficient just barely misses the traditional threshold of statistical significance). By contrast, there was no statistically significant difference between Liberal MPs and legislators from the other opposition parties about focusing on the pandemic.Footnote 6 Conservative MPs were much more likely than Liberals, however, to lead their front pages with a discussion of other policy issues and to adopt an Ottawa-centric frame. Specifically, Conservative MPs had 6.3 times the odds of Liberal MPs of leading their newsletters with a discussion of policy, and they had 6.7 times the odds for an Ottawa frame. Here again, there were no statistically significant differences between Liberal MPs and legislators from the other opposition parties. None of the other MP-level or riding-level characteristics pass the conventional threshold for statistical significance. One other coefficient that was statistically significant at the .051 level is the coefficient for cabinet ministers, who are less likely to adopt an Ottawa-centred focus.
Table 1. Determinants of Newsletter Frames March 2020–August 2021 (front page articles)
Note: Exponentiated coefficients; standard errors in parentheses
+p < .10; *p < .05; **p < .01; ***p < .001
Table 2. Predicted Probabilities of Focusing on COVID-19, Policy and Ottawa, by Partisanship
Note: Standard errors in parentheses. Other variables held at their means. Table shows the predicted probabilities from three separate models. The fourth model (leader focus) did not have a sufficiently large sample size to calculate predicted probabilities.
Table 3. Predicted Probabilities of Focusing on COVID-19, Policy, Ottawa and Party Leader, by Time Period
Note: Standard errors in parentheses. Other variables held at their means. Table shows the predicted probabilities from four separate models.
While riding characteristics and MP characteristics outside of partisanship may not have played much of a role in determining what appeared in householders, the broader context seemed more visible. MPs focused much less on the pandemic the further away from the reference period (March to June 2020) that a householder was written. They were about as likely to focus on COVID-19 from July to December of 2020 but had a much smaller probability of leading a householder with a focus on the pandemic from January to April 2021, and an even smaller probability from May to August 2021. As discussions of the pandemic receded, the topic was largely replaced with messaging around other policy issues. Newsletters had approximately 4.3 times the odds of focusing on policy in July to December 2020 compared with March to June 2020. Those odds went down to approximately 3.7 in January to April 2021, before shooting up to 9.9 for May to August 2021. Overall, the logistic regression suggests that when MPs had an opportunity to speak to their entire consistencies during this timeframe, the most important determinants of their communications strategy were their partisanship and what they perceived to be the big issues of the day—not their own personal attributes or those of their ridings.Footnote 7
Discussion and Conclusion
MPs have very few opportunities to engage directly with all their constituents. Short of knocking on doors, the only reliable, cost-efficient way for an MP in Canada to reach all households is through a printed newsletter, known as a householder, that is sent to all the addresses in their electoral district up to four times a year. MPs and their staff spend a considerable amount of time and effort preparing these documents, as they recognize them as a way to promote select information and to reach people who do not ordinarily pay attention to politics. Further, householders are a tangible opportunity to engage with constituents without drawing on the MP's office budget. Indeed, many offices receive a notable increase in constituency office traffic following each mail-out. Given the time and effort to prepare these documents and the breadth of exposure, they are likely to be a good indicator of how legislators frame their overall communications with their constituents.
In this article, we ask: What determines the topics that MPs prioritize in communications with all constituents in their electoral districts? To answer this, we examined a sample of newsletters that 145 MPs’ offices sent to all households in their ridings during the COVID-19 pandemic, and we consulted a dozen MPs’ offices for contextual information. Our quantitative data allow us to examine trends across time and MPs, while our qualitative data provide a deeper understanding of the practical considerations involved with creating householders. Moreover, examining newsletters sent during this period is useful because one issue dominated the public conversation: the pandemic itself. We can track the content of householders during this period to understand what drives how MPs choose to frame their communications with their electors: for example, with respect to householder content, whether the main issue of the day (the pandemic) was deemed to matter more or less than MPs’ own personal characteristics (such as political partisanship) or the characteristics of their riding.
While prior research suggest that MPs are free to chart their own course in their constituencies based on their own characteristics, experiences, and the needs of their communities (for example, Koop et al., Reference Koop, Bastedo and Blidook2018), our results show that partisanship is by far the most important predictor of what MPs include in their communications to their entire constituencies. By contrast, the MP's own personal characteristics and the demographics of their constituencies are much less likely to predict what MPs focus on in their communications. This suggests that while MPs see a need to balance local and national information, there is a clear prioritization of constituency matters in their overall communications strategies. MPs appear to recognize that householders are an opportunity to convey that they are actively in touch with constituents. Since MPs in marginal ridings are especially incentivized to avail themselves of householder mailings (Umit, Reference Umit2017; Spurr, Reference Spurr2015), this may also demonstrate the conditions in which mail-outs can influence Canadian public opinion and voter support, as apart from the American and British case (Cover and Brumberg, Reference Cover and Brumberg1982; Jackson, Reference Jackson2008). However, although all Canadian MPs discuss constituency matters in their householders, their party affiliation heavily influences what else they talk about.
Against a backdrop of local constituency focus, MPs talked about the COVID-19 pandemic more than any other issue during our period of study. This is unsurprising, since the pandemic was the most salient issue in Canadian public discourse at the time these documents were sent. MPs were likely driven by a desire to appear responsive to the main issue of the day, including by telling their constituents about the availability of vaccines, financial supports, or other policy decisions related to the pandemic. Through both these sets of frames, however, partisanship appeared as a constant throughline: all else equal, Conservative MPs were more likely than Liberal MPs to adopt an Ottawa-centric frame, and they were less likely than Liberal MPs to focus on the pandemic. Indeed, at the start of the COVID-19 global pandemic in March 2020, researchers noted that citizens and political representatives from a variety of partisan backgrounds largely agreed on the way forward in advanced democracies like Canada (Merkley et al., Reference Merkley, Bridgman, Loewen, Owen, Ruths and Zhilin2020). The early aftermath of the first wave of public health lockdowns was also associated with increased public reliance on information from official government sources, underscoring the importance of messaging. However, our findings show that political consensus after the start of the pandemic was short-lived. MPs were either unable or uninterested in putting politics aside in their messaging to constituents in the year after the start of public health restrictions.
The importance of partisanship was also apparent in our finding that opposition MPs were more interested in focusing on policy issues than government MPs were. This finding appears counterintuitive; as the party of government, we might expect Liberal MPs to promote their activities in government more than MPs from other parties. However, government MPs are also at greater risk of appearing out of touch with their constituents because they are preoccupied with exercising political power in Ottawa. One way for them to manage this risk is to focus on what they are doing to deliver results for their constituents. By contrast, during the study period, Conservative MPs increasingly adopted a message of standing up to Ottawa, which may be reflected in their focus on perceived policy failures on the part of the government. Indeed, Canada's Westminster-style model gives the governing party a near monopoly on the exercise of legislative power. Opposition MPs are therefore at a greater risk of appearing powerless or unable to bring about real change and may thus overcompensate by spending more of their energy talking about all the things the government is doing wrong from a policy perspective. These differences also reflect the pressures parliamentarians face from their leaders’ office, as well as the other social and political forces that promote intraparty cohesion in the Canadian Parliament (Godbout, Reference Godbout2020; Malloy, Reference Malloy2023; Marland, Reference Marland2020; Rayment and McCallion, Reference Rayment and McCallion2023). Further research is needed to establish to what extent the intraparty similarities reflect the messaging co-ordinated by the leader's office by way of each party's caucus research bureau.
What do these results mean for our understanding of the role that MPs play in Canadian representative governance? On the one hand, we know that Canadian politics is structured by very strong party discipline, elements of which MPs both lament and which can be beneficial to their political careers (Kam, Reference Kam2009; Godbout, Reference Godbout2020; Marland, Reference Marland2020). On the other hand, prior research has suggested that constituency work remains one of the last bastions of freedom that MPs have from the constraints of the central party (Koop et al., Reference Koop, Bastedo and Blidook2018) and that party election candidates have more freedom to defy party messaging than is commonly believed (Robbins-Kanter, Reference Robbins-Kanter2022; Yates, Reference Yates, Marland and Giasson2022). If this were entirely true, we would expect to find other cleavages to be more important determinants of what MPs focus on in their constituent communications—for example, things specific to the electorate's own demographic characteristics, or the MP's own attributes. Instead, our findings add nuance to this work by suggesting that in the decade since Koop et al. did their fieldwork, party discipline has caught up with at least one aspect of constituency work—communications funded by Parliament. We are not suggesting that constituent communications are an exclusively partisan affair; rather, partisanship is a key element that needs to be considered to properly understand constituent communications. This makes intuitive sense: if the leader's office was inclined to exert more influence over constituency operations, communications would seem to be a natural place to do so. Advances in technology and the professionalization of communications have enabled the centre to actively co-ordinate the distribution of templates and messaging to MPs and made it much easier to monitor MP communications for compliance.
Finally, many MPs’ offices post their householders online or will supply them if asked; however, more than half of the members of the House of Commons that we contacted repeatedly ignored or ultimately denied our request for these publicly funded newsletters. Given that some MPs and their staff are reticent to participate in academic research, scholars interested in materials issued by MPs’ offices should consider integrating crowdsourcing into their data collection efforts. Political observers need access to communications materials to establish what MPs are saying to their constituents, in order to assess similarities and differences across Canada and to consider the extent of party cohesion in representation.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0008423923000641
Acknowledgments
The authors gratefully acknowledge research assistants Abdullah Alzubaidi, Adrian Castro Carrillo, Harnoor Kochar, Colleen McCool, Gala Palavicini Jauregui and Sadaf Saeed, and thank Chris Tenove for his feedback. We also thank the MPs and their staff who provided householders and/or answered our questions, including MP Ziad Aboultaif, MP Dan Albas, MP Kody Blois, MP Nathaniel Erskine-Smith, MP Heather McPherson, MP John Nater, staffers Jennifer MacKinnon and Shawn Murphy, and staff in the offices of MPs Alexandre Boulerice and Stephanie Kusie.
Open access
